Cosmic markers, ⁴⁰Ar/³⁹Ar dating and paleomagnetism of the KT sections in Anjar area of the Deccan large igneous province

Abstract

Bhandari et al. reported the discovery of iridium-bearing sediments sandwiched between basalt flows in the Anjar area (Kutch province, India). They concluded that the signature of the K/T impact had been recorded and that onset of volcanism in the Deccan traps preceded the K/T boundary, excluding the possibility of a causal connection. This paper reports complementary analyses of Anjar outcrops by a joint Indo-French team, where we focused on cosmic markers (iridium and spinels) in the intertrappean sediments and ⁴⁰Ar/³⁹Ar dating and paleomagnetism of the lava flows. Anomalous Ir concentrations (up to 0.4 ng/g) are confirmed, with up to three thin and patchy enriched layers which cannot be traced throughout the exposed sections. Despite careful search, no Ni-rich spinels were found. Eight basalt samples provided ⁴⁰Ar/³⁹Ar results, four on plagioclase bulk samples, four on whole rocks. Spectra for whole rocks all indicate some amount of disturbance, and ages based on plagioclase bulk samples seem to be consistently more reliable. The three flows underlying the Ir-bearing sediments are dated at 66.5 Ma, and two overlying flows at 65 Ma. Magnetic analyses (both thermal and by alternating fields) uncovered clear reversed primary components in the upper flows, and more disturbed normal components in the lower flows, with evidence for an additional reversed component. There are reports that the intertrappean sediments contain uppermost Maastrichtian dinosaur and ostracod remains above the uppermost Ir-bearing level, and may not be mechanically disturbed. We propose the following scenario to interpret these multiple field and analytical observations. Deccan trap volcanism started within uppermost Maastrichtian normal chron C30N at ~66.5-67 Ma in the Anjar area. Volcanism then stopped at least locally, and lacustrine sediments were deposited over a period that could be in the order of 1-2 Ma. The K/T bolide impact was recorded as a deposit of Ir, and possibly (though not necessarily) spinels. Volcanism resumed shortly after the K/T boundary, within reversed chron C29R, as witnessed by the three reversely magnetised overlying basalt flows dated ~65 Ma. This was responsible for erosion and destruction of part of the uppermost sediments (including spinels if there were any) and heterogeneous and non-uniform redeposition of Ir at a number of underlying sedimentary levels. This was also responsible for the partial remagnetisation of Bhandari et al., and are compatible with the occurrence of the K/T impact at the paleontological K/T boundary, and of Deccan trap volcanism straddling the boundary and starting before the impact. Anjar provides evidence for minor volcanism somewhat earlier than suggested by some authors, though still within normal chron C30N. There is no indication contradicting the view that the bulk of Deccan trap volcanism occurred over only three chrons (C30N, C29R, C29N).[Courtillot, Evolutionary Catastrophes: the Science of Mass Extinctions, Cambridge University Press, 1999; Courtillot et al., Earth Planet. Sci. Lett. 80 (1986) 361-374; Vandamme et al., Rev. Geophys. 29 (1991) 159-190].